Grid tied solar question
merlin-7
higher than the grids in order to feed the grid?
Thanks
Anthony Matonak
> If you have a grid-tied solar system, does your output voltage need to be
> higher than the grids in order to feed the grid?
The voltage of the inverter has to be higher than the grid but
not by much. Since the voltage of the grid is AC and constantly
changing the inverter has to produce exactly the right voltage
and direction for that particular moment in time.
None of this is of any importance to a regular person unless
you are designing your own grid-tied inverters. All a normal
person needs to know is that it passes the UL-1741 standard
(in the United States).
Anthony
merlin-7
connected?
How could many grid-tied systems be able to function at the same time, with
all of them trying to be higher in voltage than the main grid?
Charles Quinn
news:470c27cb$0$32485$4c36...@roadrunner.com:
> So what would happen if everyone had had a grid-tied solar system
> connected?
> How could many grid-tied systems be able to function at the same time,
> with all of them trying to be higher in voltage than the main grid?
The same way all those generators do by adjusting the phase of the sine
wave.
http://en.wikipedia.org/wiki/Grid_tie_inverter
--
Charles
The significant problems we face cannot be solved
at the same level of thinking we were at when we
created them. Albert Einstein
Eeyore
merlin-7 wrote:
> So what would happen if everyone had had a grid-tied solar system
> connected?
> How could many grid-tied systems be able to function at the same time,
Plenty.
> with all of them trying to be higher in voltage than the main grid?
Are you concerned that the voltage would rise ? It would actually but
*minutely*. Same as it does now when the loading is light.
The determining factor is Ohm's Law. V= I.R.
Graham
Solar Flare
Basic physics.
"merlin-7" <merl...@sc.rr.com> wrote in message
news:470bfe5d$0$15393$4c36...@roadrunner.com...
Roderick
In practical terms, it would be hard for the grid voltage to rise too
much, because then the electric company would simply compensate by
delivering less energy. But what if every single user was generating
a lot more than they used? Yes, the grid voltage would rise, and I
think the IEEE thought of that one, too, because inverters will refuse
to operate if the grid voltage goes out of range, and that includes
too high. I guess that would kind of suck if you couldn't sell back
electricity on a sunny day because everyone else was generating too
much. I doubt it's going to happen anytime soon, though.
Another concern if everyone had grid-tied would be, what happens on a
sunny day, when everyone is generating all they need for their A/C,
then in rolls a big cloud front. Can the power company suddenly make
up that change in capacity?
Eeyore
Solar Flare wrote:
> If you are in parallel with the "grid" then the voltage is the same.
> Basic physics.
Actually, if the voltage was *exactly* the same, no current would flow.
Hence to get current to flow back into the 'grid' the inverter outputs a voltage
that's *just very slightly* greater than the 'grid' voltage.
Graham
Eeyore
Roderick wrote:
> If everyone had a grid-tied system. Interesting thought experiment.
>
> In practical terms, it would be hard for the grid voltage to rise too
> much, because then the electric company would simply compensate by
> delivering less energy.
That's *exactly* what happens ! That's the whole reason for doing it in fact.
Graham
merlin-7
"Roderick" <roderic...@lycos.com> wrote in message
news:1191991683.9...@50g2000hsm.googlegroups.com...
> If everyone had a grid-tied system. Interesting thought experiment.
>
> Another concern if everyone had grid-tied would be, what happens on a
> sunny day, when everyone is generating all they need for their A/C,
> then in rolls a big cloud front. Can the power company suddenly make
> up that change in capacity?
>
KI4ILB
Terryc
>>Another concern if everyone had grid-tied would be, what happens on a
>>sunny day, when everyone is generating all they need for their A/C,
>>then in rolls a big cloud front. Can the power company suddenly make
>>up that change in capacity?
>>
>
> A solar eclipse would be worse, I would think...
err, solar eclipses are predictable and known well in advance.
daestrom
"Eeyore" <rabbitsfriend...@hotmail.com> wrote in message
news:470C6B1F...@hotmail.com...
Actually in AC systems tied together with a line that has significant
reactance/resistance, the two ends of the line can be exactly the same RMS
voltage and you can cause power to flow in either direction by controlling
the phase angle between the two ends.
Power companies do this all the time. Ignoring the three-phase issues, for
a single phase transmission line with very low resistance to reactance
ratio...
P ~= Vs * Vr * sin(phase angle)
Where P is power flowing along line, Vs and Vr are voltages at the source
and receiver respectively and phase angle is the phase angle difference
between the two voltages. In fact, the supply voltage can actually be
*lower* than the receiving voltage on the line and power can still flow from
supply to receiver.
Grid-tie inverters can do this by simply controlling the inverter output to
lead the supply by the proper amount.
AC transmission isn't just Ohm's law.
daestrom
BobG
inverter.... 240V/100A.... a couple of ohms. If I want to pump 1KW
back into that transformer, I need to pump about 4A back up the
wire... 4A across that 2ohm load needs 8V, so if my inverter is
producing 248V out and the transformer is try to put out 240 coming
in, I'm ahead and my meter is spinning backwards. I think.
Solar Flare
115kV system via a 13.8kV system. The noise in the new switchgear was
so bad, I admit, to hiding behind brick walls on that one...LOL I just
wish I had all the recording unit screens fired up for watching the
VAR flow until the parallel was broken.
About six months later we discovered one of the Station Linemen hadn't
fastened the cable pass-through plate (split and aluminum too) bolts
down on the back of the cell and it vibrated to the circulating
currents and was making the noise via individual phase magnetic
fields.
"daestrom" <daestrom@NO_SPAM_HEREtwcny.rr.com> wrote in message
news:470d6a05$0$32556$4c36...@roadrunner.com...
Eeyore
daestrom wrote:
> "Eeyore" <rabbitsfriend...@hotmail.com> wrote
> > Solar Flare wrote:
> >
> >> If you are in parallel with the "grid" then the voltage is the same.
> >> Basic physics.
> >
> > Actually, if the voltage was *exactly* the same, no current would flow.
> >
> > Hence to get current to flow back into the 'grid' the inverter outputs a
> > voltage that's *just very slightly* greater than the 'grid' voltage.
>
>
> Actually in AC systems tied together with a line that has significant
> reactance/resistance, the two ends of the line can be exactly the same RMS
> voltage and you can cause power to flow in either direction by controlling
> the phase angle between the two ends.
>
> Power companies do this all the time. Ignoring the three-phase issues, for
> a single phase transmission line with very low resistance to reactance
> ratio...
>
> P ~= Vs * Vr * sin(phase angle)
>
> Where P is power flowing along line, Vs and Vr are voltages at the source
> and receiver respectively and phase angle is the phase angle difference
> between the two voltages. In fact, the supply voltage can actually be
> *lower* than the receiving voltage on the line and power can still flow from
> supply to receiver.
No.
> Grid-tie inverters can do this by simply controlling the inverter output to
> lead the supply by the proper amount.
>
> AC transmission isn't just Ohm's law.
Are you trying to bamboozle the OP with science ?
There are times when such a detailed explanation isn't really relevant to the
task in hand.
Graham
You
Eeyore <rabbitsfriend...@hotmail.com> wrote:
> Graham
>
No, he is explaining a Principle taught in Motor/Generator LAB in
Engineering 201 at the college level. That's why they call it
Engineering..... some folks actually did learn this stuff......
merlin-7
Let's say you have a comunity with 10,000 homes, on an island.
On this island, there is one power plant to feed electricty to those homes.
Along comes solar power and over half of homes install grid- tied solar
systems that each one produces an excess of power during the day.
How would this work?
Thanks
daestrom
"BobG" <bobga...@aol.com> wrote in message
news:1192065229.5...@g4g2000hsf.googlegroups.com...
Ah, but that transformer has a reactance as well as a resistance. Because
inductive current is 90 degrees out of phase with voltage, now you need to
lead the voltage on the other side of the transformer by a small amount as
well.....
Most distribution equipment like transmission lines and transformers have
more reactance than resistance. In a lot of power industry calcs, we just
ignore the resistance and deal only with the reactance component.
daestrom
daestrom
Care to make a wager? Been there, done it. In fact depending on the line
and the voltage support equipment at the receiving end, the voltage can
climb so far above the 'sender' that you have a problem with substation
voltage going too high out of spec, even when the generator regulator is
working just fine maintaining the 'sender' voltage.
(hint, don't connect the two voltages with a resistor alone, use an
inductance such as the pole transformer or a transmission line)
See chapter 11 of 'Handbook of Electric Power Calculations'
or,
Chapters 3 through 6 of 'Electric Power Transmission Systems'
>
>> Grid-tie inverters can do this by simply controlling the inverter output
>> to
>> lead the supply by the proper amount.
>>
>> AC transmission isn't just Ohm's law.
>
> Are you trying to bamboozle the OP with science ?
No, just trying to correct the erroneous statement that you made and set the
record straight.
>
> There are times when such a detailed explanation isn't really relevant to
> the
> task in hand.
>
Then you shouldn't make statements like...
">> > Actually, if the voltage was *exactly* the same, no current would
flow."
which is patently false. If you don't know that much about AC power
transmission, you shouldn't be making such statements.
I pointed out you were wrong and explained just what does happen in AC power
transmission. Yes, it is a bit off topic, but hardly a monstrous breach of
protocol.
daestrom
daestrom
Well, your 'thought experiment' is exactly the sort of thing that some of us
in other news groups have been discussing. As long as the grid-tie output
is a small fraction of the total grid generation, like it is now in the US,
there isn't really any problem at all.
But if grid-tie inverter-supplied power were to grow to a truely large
percentage of total generation, such as you put forth in your hypothetical
situation, then there would be problems with stability and control. All the
individual grid-tie inverters would presumably be pumping power back into
the grid and the one power plant would be 'throttling back' to try and
maintain system frequency. If that single unit can regulate fast enough
when the clouds pass overhead and then clear, system frequency could be
maintained. Of course, if the variation in frequency caused by these sudden
changes is too wide, then the inverters might decide that the grid is lost
momentarily and shut down. This in turn would suddenly dump all the load
onto the power plant. Assuming it regulates well and restores grid
parameters, then the various grid-tie inverters would see a 'restoration' of
grid power and start back in sending power to the grid. This may or may not
cascade again and again, depending on sensitivity of the inverters, the
responsiveness of the power plant and the details of the distribution
network. It might work fine, but it might get real ugly, real fast.
And of course if the total inverter output is greater than the total load,
then the excess power will start motoring the power plant and raising system
frequency. The power plant will trip on 'reverse power' and the frequency
will become very unstable. Once the power plant trips, the frequency will
swing around and the inverters will shut down thinking the grid is lost.
But this time, it really will be lost and the power plant is off-line, so
everybody 'goes dark'.
A second problem with such a system is the matter of voltage regulation.
When the power is flowing in one direction, from plant to substation to
homes, all the voltages may be in spec. But if a lot of inverters started
power flowing from homes to a substation, across the island to other
substation to homes, then voltage at various points might not stay within
spec. The 'generating' homes may see a rise in their local voltage. This
depends a lot on the power factor of the loads and the various reactances in
the distribution system. Even if the power plant generator is maintaining
its voltage, there could be areas seeing over/under voltage conditions.
If some substations are set up with tap-changing transformers or other
voltage regulating equipment, when the power flow is reversed the
'regulating action' can be wrong. For example: A tap-changing transformer
is normally set such that when the LV side rises too high it steps to
increase the turns ratio and thus 'step-down' the HV power further to a
lower LV voltage. But if the power is actually flowing in the other
direction and the LV side has a high voltage because a lot of grid-tie units
are forcing power 'upstream', then when it 'steps' to a higher turns ratio
it will actually be raising the HV side voltage even higher. And it may not
appreciably change the LV voltage so it may step repeatedly until it reaches
the end of its travel, resulting in a very high HV voltage. But it would
cause the substation to take all the reactive loading away from the power
plant and cause problems there.
These are at least some of the reasons why many utilities have an upper
limit on the total amount of grid-tie solar they are willing to connect. To
connect more would require more study and could risk the quality of service
to neighbors.
But we've got a long way to go before we get to this point except on your
hypothetical island.
daestrom
Roderick
It would have been easier to do this puzzle if you said all the homes
were producing a surplus. Since it's just half, my thoughts:
- The surplus electricity is probably not enough to run the rest of
the community. If it is, then the homeowners are very atypical (i.e.,
they don't run their air conditioners, while everyone else does)
- The power company must keep a generator running, to handle
fluctuations in demand and generation. Even if Solar is generating
all the power, the utility still must keep the generator running, to
avoid startup time. None of this applies if this area does not have a
cultural expectation of always-on electricity (places like Iraq). But
the community is wealthy enough to install all this solar, so I assume
they want power all the time. Besides, the power company provides the
50/60 Hz standard that all the grid-tied inverters must synchronize
to.
- The power company is going to have all kinds of challenges, if every
day, when the sun goes down, people go home and cook their dinners in
electric ovens, etc. They would not run efficiently, and probably
would have fought tooth and nail to block so many people from getting
solar.
- The scenario of line voltage rising to an unreasonable level is
still possible if everyone is producting way too much. The final
safeguard is that inverters would shut off - the most sensitive ones
first. However, with that sort of scenario looming, the power company
might now have the job of inserting dummy loads onto the system to
keep the voltage in range. But nobody wants to just dump all that
extra energy, so maybe they could make ice, or smelt iron, refine
silicon, or electrolyze water. Frankly, I'm not too worried about the
"too much energy" case. Humans are very good at using up energy if
it's there.
Solar Flare
"ferroresonance" by some.
If you use capacity to power factor correct the end of a transmission
line the current will drop and the voltage will rise. This will
continue until the receiving end has a higher voltage than the supply
end. Think of the inductive transmission line and tha capacitance at
the load end as a tuned ressonance circuit, If the resistive load is
removed the high voltage created can blow HV equipment up like
lightning.
"daestrom" <daestrom@NO_SPAM_HEREtwcny.rr.com> wrote in message
news:470eafaa$0$26344$4c36...@roadrunner.com...
Eeyore
You wrote:
> Eeyore <rabbitsfriend...@hotmail.com> wrote:
> >
> > Are you trying to bamboozle the OP with science ?
>
> No, he is explaining a Principle taught in Motor/Generator LAB in
> Engineering 201 at the college level. That's why they call it
> Engineering..... some folks actually did learn this stuff......
And you think the OP is capable of understanding this ? Or is it merely for 'show
off' reasons ?
Graham
Eeyore
daestrom wrote:
> "BobG" <bobga...@aol.com> wrote
>
> > The grid looks like the transformer on the pole to my grid tie
> > inverter.... 240V/100A.... a couple of ohms. If I want to pump 1KW
> > back into that transformer, I need to pump about 4A back up the
> > wire... 4A across that 2ohm load needs 8V, so if my inverter is
> > producing 248V out and the transformer is try to put out 240 coming
> > in, I'm ahead and my meter is spinning backwards. I think.
>
> Ah, but that transformer has a reactance as well as a resistance. Because
> inductive current is 90 degrees out of phase with voltage, now you need to
> lead the voltage on the other side of the transformer by a small amount as
> well.....
Your idea that transformers present significant reactance to the load/source is
plain WRONG.
Read up on 'leakage inductance'.
Graham
Eeyore
daestrom wrote:
> "Eeyore" <rabbitsfriend...@hotmail.com> wrote in message
>
> Then you shouldn't make statements like...
>
> ">> > Actually, if the voltage was *exactly* the same, no current would
> flow."
>
> which is patently false.
It's *** 100% *** correct in the instance we're discussing where an inverter is
phase locked to the incoming supply.
Graham
nicks...@ece.villanova.edu
>"merlin-7" <merl...@sc.rr.com> wrote in message
>> Let's say you have a comunity with 10,000 homes, on an island.
>>
>> On this island, there is one power plant to feed electricty to those homes.
>>
>> Along comes solar power and over half of homes install grid- tied solar
>> systems that each one produces an excess of power during the day.
>>
>> How would this work?
>
>... As long as the grid-tie output is a small fraction of the total grid
>generation, like it is now in the US, there isn't really any problem at all.
That makes things simpler.
>But if grid-tie inverter-supplied power were to grow to a truely large
>percentage of total generation, such as you put forth in your hypothetical
>situation, then there would be problems with stability and control.
And different solutions.
>All the individual grid-tie inverters would presumably be pumping power
>back into the grid and the one power plant would be 'throttling back'
>to try and maintain system frequency. If that single unit can regulate
>fast enough when the clouds pass overhead and then clear, system frequency
>could be maintained. Of course, if the variation in frequency caused by
>these sudden changes is too wide, then the inverters might decide that
>the grid is lost momentarily and shut down.
In another scenario, the "grid-tie inverters" get precise synchronization
and power control information from the power plant and adjust themselves
to its requirements, in a sort of supply vs demand management.
There's an interesting Dutch variation on pumped storage (on paper): build
up the walls of an offshore underwater bowl and pump seawater out to store
energy, and let it flow back in to retrieve energy. A floating version might
be interesting, maybe a part of a floating city...
Nick
daestrom
There's a difference between 'significant reactance' (i.e. what you
*thought* I said) and the ratio between reactance and resistance. Most of
the impedance that transformers present to the load/source is reactance, not
resistance. If I had meant that it was 'significant' the way you interpret,
I wouldn't have said "...you need to lead the voltage on the other side of
the transformer by a *SMALL* amount as well" [emphasis added].
So when you calculate the four-terminal network solution for AC power
transmission through a line with a transformer, you'll find the transformer
adds almost no resistance but it does add reactance term that needs to be
accounted for. Reactance of a small transformer such as on a utility pole
can be between 3% and 5% (you do understand how P.U. calculations are used
to express the reactance of a power transformer in percent don't you?) Or
why don't you just admit you've never done any power transmission
calculations and go away?
daestrom
merlin-7
solar systems that are tied together.
To help me understand this, lets take this up a notch.
Let's say that we have several homes with grid tied systems but the homes
solar systems provide all the electricty, except for one generator that
supplys the signal current to the grid, to get the ball rolling.
Lets ignore nights or no sun for this.
Could it work?
In a nut shell, what would happen (to the grid) If A way to make PV panels
suddenly became very cheap and everyone could buy a solar PV system that
produces more power than they use?
Lets leave out the part where the government and corporations squash it
....
Thanks
Anthony Matonak
...
> Let's say that we have several homes with grid tied systems but the homes
> solar systems provide all the electricty, except for one generator that
> supplys the signal current to the grid, to get the ball rolling.
>
> Lets ignore nights or no sun for this.
> Could it work?
The way the grid and grid-tied PV systems are put together now, No.
Could it be made to work? Certainly. You would need more intelligent
grid-tied inverters, a central clock that synchronizes them all and
some way to deal with times when there is more supply than demand.
One method would be diversion or deployable loads. These would be
things that consume lots of electricity but aren't very picky about
when or how much.
Another method would be some way to store the energy. This could store
the excess and then put it back on the grid at times when there is too
little supply.
Another method would be some form of automatic control that reduces
the amount of energy each (or some) inverters feed back into the grid.
Anthony
Eeyore
daestrom wrote:
> "Eeyore" wrote
> > daestrom wrote:
> >> "BobG" wrote
> >>
> >> > The grid looks like the transformer on the pole to my grid tie
> >> > inverter.... 240V/100A.... a couple of ohms. If I want to pump 1KW
> >> > back into that transformer, I need to pump about 4A back up the
> >> > wire... 4A across that 2ohm load needs 8V, so if my inverter is
> >> > producing 248V out and the transformer is try to put out 240 coming
> >> > in, I'm ahead and my meter is spinning backwards. I think.
> >>
> >> Ah, but that transformer has a reactance as well as a resistance.
> >> Because inductive current is 90 degrees out of phase with voltage, now you
> need
> >> to lead the voltage on the other side of the transformer by a small amount
> >> as well.....
> >
> > Your idea that transformers present significant reactance to the
> > load/source is plain WRONG.
> >
> > Read up on 'leakage inductance'.
>
> There's a difference between 'significant reactance' (i.e. what you
> *thought* I said) and the ratio between reactance and resistance. Most of
> the impedance that transformers present to the load/source is reactance, not
> resistance.
Nonsense.
A 'perfect' transformer would present only the reflected impedance from the
primary. To that has to be added the leakage inductance which in a good design
should be very low.
The idea that transformers are highly reactive is a complete myth.
Graham
David Williams
-> suddenly became very cheap and everyone could buy a solar PV system that
-> produces more power than they use?
It would be fairly easy if direct feeding of power into the grid were
forbidden. Each person would generate as much as he needs, and maybe
put the surplus into batteries. When the sun doesn't shine, he can
draw on his battery reserve, and maybe take power from the grid, which
would still be there as a backup. If he finds himself with a whole lot
of charged batteries, he could take them to the power company, which
would use them at its own convenience to add power to the grid. Of
course, the company would pay to do this.
So there wouldn't be the problem of people pushing power backwards into
the grid.
dow
Bill Kaszeta / Photovoltaic Resources
for grid tied inverters may years ago. It is call 'islanding', an island of
interconnected PV inverters and distributed loads.
Inverters meeting the existing standards (IEEE, IEC, etc.) will detect islanding
and shut down. This was tested with small islands and can be readly
modeled with computer simulations. Decaeds ago there were many technical
papers on the subject.
Bill Kaszeta
Photovoltaic Resources Int'l
Tempe Arizona USA
bi...@pvri-removethis.biz
David Williams
-> > resistance.
-> Nonsense.
-> A 'perfect' transformer would present only the reflected impedance from the
-> primary. To that has to be added the leakage inductance which in a good desi
-> should be very low.
-> The idea that transformers are highly reactive is a complete myth.
-> Graham
Since when are real transformers "perfect".
If there is no load, i.e. an open circuit on the secondary, then the
primary is purely inductive.
dow
Eeyore
David Williams wrote:
The current flow in the primary is purely (well mostly at least) inductive under no
load conditions. That doesn't affect the impedance seen at the secondary though.
Graham
You
david.w...@bayman.org (David Williams) wrote:
> If he finds himself with a whole lot
> of charged batteries, he could take them to the power company, which
> would use them at its own convenience to add power to the grid. Of
> course, the company would pay to do this.
>
I can see it now, Pickup Trucks with a couple of Tons of Batteries in the
back, driving down to the local PUD, (Public Utility District) and
connecting to a GIANT Inverter with BIG Cables. The PUD then sucks the
batteries dry, while you stand around and have a Starbucks. Then off you
go hauling that 2 Tons of Lead back home, and reinstall them in your
BatteryBox.....
Yea, Right............
R.H. Allen
So is the idea that anybody SAID they are highly reactive.
As daestrom has already tried to explain, there's a HUGE difference
between "most of the impedance ... is reactance" (what daestrom said)
and "transformers are highly reactive" (what you seem to think he said).
On top of that, he noted that utility pole transformers have reactances
between 3% and 5% -- how on earth did you manage to interpret that as
"highly reactive"?
Mel
Probably one of the simplest ways to use power that is deemed not
acceptable for the security of the grid (for whatever reason) is an
electrical resistance in a tank of water... much easier than managing
batteries.
Mel
reinforcement even for relatively small peak powers.
I've had a few cases where farmers wanted to install 10 to 20kWp systems
on their buildings, and abandoned the project because the grid manager
required them to pay for a now distribution post / new bigger section
cables (sorry if I use the wrong vocab,hope you'll be able to follow
anyway).
On their local line there are generally only a few customers (3 to 10)
with connections in the 12 to 36kVA range. When the line was installed,
the grid manager designed the line to run at the upper extremes of the
the operating limit at the distribution post, and the oppposite (lower)
extremes at the users delivery point - this allowed them to use smaller
section wires basically.
Now adding a pv system in there introduces the risk that on sunny summer
days, when consumption is at a low (not much air con around here) and
production peaks, the electricity characteristics will go over the
extreme limits at the distribution point, even though they are
completely within the operating range at the delivery point.
So end-of-line producers are now having to pay for the grid managers
economic shortcuts made in the past...
And that's just adding to the fact that those who invested in small
systems under 5kWp have a frustrating time because the inverter
operating range (tension / frequency) is stricter than the grid
operation range, and over here the grid manager will only accept factory
modifications to inverter range characteristics, but inverter
manufacturers will not do any factory adjustments, and insists that
installers or the grid manager do it. And of course, those on the end of
the line pretty much always have electricity at the extremes of the grid
managers operating range.
I wonder if we could convince the grid manager to design all new grid
extensions as loops? (I know, I'm dreaming!)
Mel
merlin-7 a écrit :